Some notes I jotted down in response to JB’s treatise on the stream
JB’s Post: http://bit.ly/eAL2Z
Is 15-20% traffic via social distribution a new phenomenon or just a phenomenon that is newly trackable? Perhaps social distribution was always a traffic driver, but it used happen via email, instant messenger, and harder to track channels? Even an organic google search for omgpop leading to a hit on omgpop.com might classify as social distribution—word of mouth is probably the true driver of many such searches.
Love prezi ;)
It seems one of the points John is driving towards is real time as the solution to overflow, “now” as the filter. Ideally, I want a contextual filter for the deluge of data in my stream, but the “now” filter is an acceptable substitute, I think. If this is to be the filter, I think AC’s “persistence” score will be an important complementary filter to fill me in on the important bits I missed when I was not tuned in to the “now” filter.
Disqus needs to go realtime—I just thought this and then read Fred’s comment and realized that email is quite near real time.
“the one thing that you really didn’t weave into this is the role of mobile.
twitter’s shortcode is the most used shortcode in the US. that’s a telling piece of data.
this real time stream is with us all the time, and it would not happen without mobility” — important point Fred makes here. I wonder how much of the solution is hardware, how much is mobile data infrastructure (towers and such), how much is software? I’m sure they are all important, but are the infrastructure and devices sufficient yet?
via http://bit.ly/eAL2Z
Tumblr Search / SEO - ARRGG!
Tonight I was trying to find a link to this post on my blog of a great cover of Hey Jude: http://bit.ly/aG4H
It took me 2 failed searches using Tumblr search ( blog.kortina.net/search/term ), 2 failed Google searches, and a third Google search that finally led me to Disqus where after 2 pages I was able to locate the post.
If I can’t even find things I know that I posted here, how is anyone supposed to discover a post? I realize discovery on Tumblr is supposed to happen through the dashboard, organically, but I’m not sure it’s worth the trade if it means I can’t find my old posts. Going to see what’s going on with my Wordpress blog.
“Twitter presents their audience with a challenge - if you don’t like what we’ve done, you have every means to make it better for free, so don’t complain, do something. The creation and use of platforms is an admission that no company has a lock on innovation, but we all can benefit quickly by aligning ourselves with the innovation stack.”
Greg Battle.
More Greg: “Twitter chose wisely in that they have used the platform to export all interface innovation (the most mutable and therefore, costly part of every web application).”
Had not thought about this angle, but really insightful. In a way, I would say MySpace has done the same. Facebook has not.
(via betaworks)
Great point by Greg, agreed.
Have to disagree about Facebook—they are taking this “outsourcing” to an even greater extreme than Twitter or Myspace with FB Connect, with which they are essentially just giving away the social graph and privacy controls they have built. Where they are headed is beyond Myspace: instead of having to go to a URL on the host domain, you can go to any app (with any design, any UI) and see the full names, faces, and actions of your friends.
This means, if you wanted to, you could build a full Twitter or Myspace that used Facebook Connect for the relationship / friendship model. Quite crazy and an interesting strategy, considering that this social graph is the source of all of Facebook’s value. Umair would approve of the radical openness, I think. I wonder if Rafer’s premonitions about imminent lockdown of connect will hold? I just don’t see it happening, given that is is not a spam channel, to be exploited like News Feed was, but rather a leasing of contextual data.
Another important part of the “lease” granted by Twitter, FB, and Myspace is the namespace, which I was discussing Andy earlier today. I wonder whether the Twitter / Myspace model is better than Facebook. With Twitter and Myspace, there is a global namespace, meaning that searching “kortina” will return the same thing for anyone conducting this search. Terms map one-to-one with entities.
As Navajeet recently pointed out to me, Facebook is the only company with search intelligent enough to show me my friend when I type “Jason Omara” vs. the popular actor with the same name. The global namespace seems more lucrative to the person in control of the namespace, but the contextual namespace is far more interesting.
